wizard
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by reading repository files to scope the wizard.
- Ingestion points: Reads .env, .env.example, README, and .github/workflows/* to identify necessary environment variables.
- Boundary markers: No delimiters or safety instructions are present to mitigate malicious content in ingested files.
- Capability inventory: Generates bash scripts with capabilities for URL opening, file writing, and GitHub secret management.
- Sanitization: No content validation or escaping is applied to the ingested data before script generation.
- [CREDENTIALS_UNSAFE]: The skill is designed for credential management. It reads local .env files to provide defaults and generates scripts that store secrets in .env or transmit them to GitHub Actions via the gh CLI.
- [COMMAND_EXECUTION]: The skill generates executable shell scripts and utilizes the gh CLI and system tools like xdg-open for its operations.
Audit Metadata